106 teams. 16 locations. See this year's winner.

106 teams. 16 locations. Almost 600 UBS employees got together for the fifth edition of our Hackathon. Colleagues with all different professional backgrounds, from developers to UX designers, project managers to marketing gurus, pitched innovative ideas and collaborated to bring them to life.

Why do we do it? Collaboration is the core of UBS, it's in our DNA. This is why, every year, our Hackathon gets bigger and better. It's an opportunity for colleagues to get engaged, collaborate across functions and also for some good out-of-the-box thinking and to give a unique contribution in innovating our company. Together.

This was our most global Hackathon yet. Our locations in Weehawken, Nashville, Stamford, Chicago, Sao Paulo, Tokyo, Shanghai, Hong Kong, Singapore, Sydney, Pune, Mumbai, London, Zurich, Krakow and Wroclaw were buzzing for two days with brilliant ideas!

Of course, to add to the electricity, we had to throw in a little friendly competition. The ideas with the best scores in use of technologies, business impact, creativity and innovation, viability and design became the winners for each region across the globe and will even get the chance to become reality within the company.

Curious? Discover the winning ideas below.

Team name: Team Factory

The project: Using the Apple speech API (Siri) to automate user commands and fetch client account balances without using the client mobile application.

Technology used: Apple Speech API (Siri).

Team name: Data Hunters

The project: ‘E.Z. Mail’, a non-supervised machine learning solution, which categorizes incoming emails according to the user's preferences. It judges email priority from topic keywords and user actions history. It learns continuously and allows users to drive the learning, helping your mailbox stay focused.

Technology used: Machine learning and nature language processing including word2vec, nonnegative matrix factorization (NMF), latent Dirichlet allocation (LDA) models and Python win32com to interact with Outlook.

Team name: Advisory Stuff

The project: ‘Advisory Stuff’ allows a client (or their advisor) to health check problems in a portfolio ,in real-time, by utilizing cloud-based machine learning technology .

Technology used: Google Cloud (with a self-built backend), Python (Flask), data encryption, machine learning (using a self-built algorithm), Angular 2, primeNg and SQLite database.

Team name: Limitless

The project: ‘Limitless’ collates data from several sources to provide a comprehensive view of relevant market movements for users to analyze.

Technology used: Bootstrap, amcharts, angular, Java, OpenNLP, Spring Boot, ROME for RSS feed processing and the twitter API.

Are you interested in joining our team?

Visit our job board and you may find something perfect for you.

Enjoyed reading this? You might also like…